Finite-dimensional compensators for the H∞-optimal control of infinite-dimensional system via a Galerkin-type approximation

Ming Qing Xiao, Tamer Basar

Research output: Contribution to journalConference articlepeer-review

Abstract

We study the existence of general finite-dimensional compensators in connection with the H-optimal control of linear time-invariant systems on a Hilbert space with noisy output feedback. The approach adopted uses a Galerkin-type approximation, where there is no requirement for the system operator to have a complete set of eigenvectors. We show that if there exists an infinite-dimensional compensator delivering a specific level of attenuation, then a finite-dimensional compensator exists and achieves the same level of disturbance attenuation. In this connection, we provide a complete analysis of the approximation of infinite-dimensional generalized Riccati equations by a sequence of finite-dimensional Riccati equations. As an illustration of the theory developed here, we provide a general procedure for constructing finite-dimensional compensators for robust control of flexible structures.

Original languageEnglish (US)
Pages (from-to)1095-1100
Number of pages6
JournalProceedings of the IEEE Conference on Decision and Control
Volume2
StatePublished - 1999
Externally publishedYes
EventThe 38th IEEE Conference on Decision and Control (CDC) - Phoenix, AZ, USA
Duration: Dec 7 1999Dec 10 1999

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Modeling and Simulation
  • Control and Optimization

Fingerprint

Dive into the research topics of 'Finite-dimensional compensators for the H∞-optimal control of infinite-dimensional system via a Galerkin-type approximation'. Together they form a unique fingerprint.

Cite this